| Time |
Nick |
Message |
| 00:11 |
bshum |
Hmm, something to ponder tomorrow |
| 00:11 |
bshum |
When I pull from https://code.launchpad.net/~denials/evergreen/translation-export |
| 00:12 |
bshum |
And I look at the actual po file name from say... build/i18n/po/tpac |
| 00:12 |
bshum |
I see it's called cs.po |
| 00:13 |
bshum |
Not cs-CZ.po |
| 00:13 |
bshum |
Which is what the file seems to be named in master |
| 00:13 |
bshum |
Could that be why some strings don't translate because the process we're using is not comparing the files appropriately based on the different filenames? |
| 00:13 |
bshum |
And so we're not updating i18n from LP correctly |
| 00:14 |
* bshum |
adds this to his list of things to figure out later. |
| 00:35 |
dbs |
@later tell bshum I thought I had a script to map the names appropriately, but it's been a long time... |
| 00:35 |
pinesol_green |
dbs: The operation succeeded. |
| 00:38 |
dbs |
@later tell bshum yeah, build/i18/scripts/update_pofiles |
| 00:38 |
pinesol_green |
dbs: The operation succeeded. |
| 00:42 |
dbs |
@later tell bshum Documented here: http://docs.evergreen-ils.org/2.6/_updating_the_translations.html |
| 00:42 |
pinesol_green |
dbs: The operation succeeded. |
| 00:53 |
|
cherri joined #evergreen |
| 01:45 |
|
dbwells_ joined #evergreen |
| 01:51 |
bshum |
dbs++ # thanks for the laters. I wonder about tpac then. It's not reflected in the list of pofiles |
| 01:52 |
|
RoganH joined #evergreen |
| 01:54 |
bshum |
There's a few directories in build/i18n/po that differ from master and the script |
| 02:20 |
bshum |
Oh yeah... adding those extra directories to the script really changes the diff for po using update_pofiles. |
| 02:20 |
* bshum |
puts it all into a new working branch for further study |
| 02:26 |
bshum |
Maybe: https://git.biblio.org/?p=bshum/Evergreen.git;a=shortlog;h=refs/heads/user/bshum/update-pofiles-script |
| 02:26 |
bshum |
Oops, wrong system :) |
| 02:27 |
bshum |
I meant: http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/bshum/update-pofiles-script |
| 02:28 |
bshum |
Which is the working branch with changes to update_pofiles and then I ran the new changed script to get new PO and make newpot again. We might backpeddle some of that as metadata only changes actually... |
| 02:28 |
bshum |
But we were talking about whether we care or not about that sort of churn/noise. |
| 02:28 |
bshum |
Something to poke at more later today. |
| 02:36 |
bshum |
Force pushed with a revised newpot commit without too much of the metadata changes |
| 03:07 |
|
cherri joined #evergreen |
| 04:13 |
|
cherri joined #evergreen |
| 05:45 |
pinesol_green |
Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html> |
| 05:49 |
|
dreuther_ joined #evergreen |
| 07:58 |
|
eeevil joined #evergreen |
| 07:58 |
|
mtate joined #evergreen |
| 07:59 |
|
phasefx joined #evergreen |
| 07:59 |
|
Callender joined #evergreen |
| 08:00 |
|
DPearl joined #evergreen |
| 08:13 |
|
rjackson-isl joined #evergreen |
| 08:18 |
eeevil |
Hack-a-way folken: breakfast? Yes! Shall we wait a couple minutes for anyone? |
| 08:21 |
eeevil |
UPDATE: 8:30-ish, eh? EH? |
| 08:21 |
eeevil |
The Yolk cafe |
| 08:24 |
|
tspindler joined #evergreen |
| 08:28 |
|
Dyrcona joined #evergreen |
| 08:32 |
|
mrpeters joined #evergreen |
| 08:40 |
|
mmorgan joined #evergreen |
| 08:42 |
|
cherri joined #evergreen |
| 08:45 |
|
kmlussier joined #evergreen |
| 08:47 |
|
Shae joined #evergreen |
| 08:47 |
|
vanya joined #evergreen |
| 08:49 |
|
ericar joined #evergreen |
| 08:49 |
vanya |
Hello everyone! |
| 08:49 |
kmlussier |
vanya: Hello! |
| 08:49 |
vanya |
Can someone tell me how patches can be submitted for evergreen? |
| 08:49 |
vanya |
Hi kmlussier ! :D |
| 08:50 |
kmlussier |
vanya: We use git to submit patches. http://evergreen-ils.org/dokuwiki/doku.php?id=dev:git provides a pretty good overview of how to submit a patch using git. |
| 08:50 |
jeff |
~contributing |
| 08:51 |
Dyrcona |
jeff: Say, what? |
| 08:51 |
kmlussier |
pinesol_green: Are you awake? |
| 08:51 |
pinesol_green |
kmlussier: It reads like a Nigerian 419 scam, but I think it is a sincere question sent to the wrong list. |
| 08:51 |
jeff |
vanya: that gives a bit of an overview -- there's also http://evergreen-ils.org/dokuwiki/doku.php?id=contributing |
| 08:51 |
jeff |
~contributors |
| 08:52 |
jeff |
huh. guess neither of those factoids exist in pinesol_green |
| 08:52 |
vanya |
Hello jeff ! :D Thank you- I'll check it out |
| 08:52 |
Dyrcona |
You'll just have to add them. |
| 08:52 |
jeff |
yeah, after i search to see if it's there under another name. |
| 08:53 |
vanya |
While going over the bugs, I noticed a bug related to the the installation of openSRF (https://bugs.launchpad.net/evergreen/+bug/801639) |
| 08:53 |
pinesol_green |
Launchpad bug 801639 in Evergreen "configure of --enable-python succeeds even though OpenSRF Python is absent" (affected: 1, heat: 6) [Wishlist,Triaged] |
| 08:53 |
vanya |
It was tagged- "triaged" . What exactly does that mean? |
| 08:54 |
kmlussier |
vanya: It just means that somebody has look at it, but nobody is working on it. |
| 08:55 |
kmlussier |
vanya: This might help you as you look through bugs - http://evergreen-ils.org/dokuwiki/doku.php?id=dev:bug_wrangler:faq |
| 08:56 |
|
cherri joined #evergreen |
| 09:02 |
|
yboston joined #evergreen |
| 09:04 |
|
jwoodard joined #evergreen |
| 09:17 |
kmlussier |
yboston: Are you setting LP bugs to fixed released as you commit snigdha's contributions? |
| 09:18 |
kmlussier |
Or, better yet, fix released. |
| 09:18 |
yboston |
no, I am not being very organized since I am multitasking :( |
| 09:19 |
kmlussier |
yboston: Sounds like the story of my life. ;) |
| 09:19 |
yboston |
I had forgotten to do that, though I had been trying to close bugs in Github, but forgot about LP |
| 09:22 |
|
jwoodard_ joined #evergreen |
| 09:23 |
|
jwoodard joined #evergreen |
| 09:36 |
|
cherri joined #evergreen |
| 09:44 |
|
DPearl joined #evergreen |
| 09:47 |
|
berick joined #evergreen |
| 09:48 |
|
Hackaway joined #evergreen |
| 09:48 |
|
vanya joined #evergreen |
| 09:48 |
Hackaway |
https://plus.google.com/hangouts/_/g7zelvxmqsbpe34kymwfzzckhqa |
| 09:49 |
kmlussier |
Hackaway: I may be running a little late today. Need to prepare for tomorrow's meeting with my Board. |
| 09:49 |
* kmlussier |
would much rather be hacking. :) |
| 09:50 |
Dyrcona |
kmlussier: I think there might be a bug with standing penalties and the current negative balance branch. |
| 09:51 |
kmlussier |
Dyrcona: I was planning to look at the generation of new overdues with that branch today. Anything else I should look at while I'm there? |
| 09:51 |
Dyrcona |
kmlussier: I see single patrons with 3 entries for max fines. Of course, I discovered this while testing a script to recalculate penalties so those entries are now gone from my database. |
| 09:52 |
Dyrcona |
kmlussier: I'll show you. |
| 09:53 |
|
RoganH joined #evergreen |
| 10:01 |
|
vanya joined #evergreen |
| 10:01 |
RoganH |
Right now everyone is checking email and settling in but we were discussing starting to tackle somethings on the wiki again about 10:15ish. |
| 10:03 |
|
yboston joined #evergreen |
| 10:04 |
|
vanya joined #evergreen |
| 10:13 |
dbs |
bshum++ # yep, updating update_pofile makes a ton of sense |
| 10:14 |
|
jboyer-laptaupe joined #evergreen |
| 10:14 |
eeevil |
dbs: 1) can I toss you a key to toss into the gitolite repo for "miker"? and 2) assuming so, where would you like me to throw it? |
| 10:15 |
bshum |
dbs: Thanks for showing me the way :) |
| 10:15 |
bshum |
dbs++ |
| 10:17 |
dbs |
eeevil: 1) yes 2) anywhere - pm works |
| 10:17 |
eeevil |
dbs++ |
| 10:18 |
eeevil |
yay, fetch! |
| 10:18 |
dbs |
eeevil: you should now have OpenSRF/Evergreen working repo commit access |
| 10:19 |
dbs |
I guess you probably want origin commit access too, eh? |
| 10:19 |
dbs |
(assuming you are the real "eeevil'' -- hmm0 |
| 10:20 |
eeevil |
YOU'LL NEVER KNOW! |
| 10:23 |
kmlussier |
For negative balances, I was just wondering what our prospects are for getting it into 2.next and what I can do to help. |
| 10:24 |
kmlussier |
I hate hearing my voice talk back to me. |
| 10:26 |
kmlussier |
I'm having trouble hearing dbwells. |
| 10:28 |
kmlussier |
I think we've conceded on the point of using voids. But my question is how we move the current branch forward. |
| 10:29 |
bshum |
kmlussier: Right, I'm going to mention that. :) |
| 10:32 |
|
_bott_ joined #evergreen |
| 10:33 |
bshum |
Discussion on https://bugs.launchpad.net/evergreen/+bug/1198465 |
| 10:33 |
pinesol_green |
Launchpad bug 1198465 in Evergreen "Support for Conditional Negative Balances" (affected: 14, heat: 62) [Wishlist,Confirmed] |
| 10:33 |
kmlussier |
In most cases, we see that the branch is behaving as it should. But we're still having problems with historical transactinos. |
| 10:35 |
csharp |
https://docs.google.com/document/d/1lL9S07c8X_zfXSexXfnwzRllnvVvJsW0qf65QxGtH7A/edit - editable hackaway doc link |
| 10:36 |
bshum |
We're now talking about https://bugs.launchpad.net/evergreen/+bug/1373693 |
| 10:36 |
pinesol_green |
Launchpad bug 1373693 in Evergreen "call number browse at top org_unit extremely slow" (affected: 1, heat: 6) [Undecided,New] |
| 10:41 |
|
dbwells joined #evergreen |
| 10:45 |
|
mnsri joined #evergreen |
| 10:55 |
jboyer-laptaupe |
berick: Finally got it up here: collab/jboyer/lp1210541_del_acpl |
| 10:58 |
berick |
jboyer-laptaupe++ |
| 11:02 |
|
vlewis joined #evergreen |
| 11:03 |
berick |
jboyer-laptaupe: tiny addition pushed |
| 11:03 |
berick |
that will cover any autofieldwiget-generated copy loc lists, common in ACQ |
| 11:05 |
jboyer-laptaupe |
berick++ |
| 11:05 |
jboyer-laptaupe |
Now back to my flailing about semi-randomly. :D |
| 11:05 |
berick |
jboyer-laptaupe: suggest focusing next on Application/Circ/CopyLocations.pm |
| 11:06 |
berick |
i believe that's the main access point for most clients |
| 11:06 |
berick |
it's mainly cstore calls |
| 11:06 |
bshum |
jboyer-laptaupe: You also noticed that z3950_index_field_map doesn't restore properly right? |
| 11:06 |
bshum |
During pg_restore I mean |
| 11:06 |
jboyer-laptaupe |
OH, hey. That is almost certainly related to my issue. I thought acpl's were so basic that the Storage/Publisher/asset.pm was the only reference to them. Maybe this isn't actually that difficult. |
| 11:07 |
jboyer-laptaupe |
bshum: yeah, depending on the phase of the moon and the -j number you through at pg_restore. |
| 11:22 |
Dyrcona |
bshum: On z3950_index_field, I've noticed on 9.1 that sometimes it restores correctly, sometimes it doesn't. |
| 11:24 |
bshum |
Dyrcona: we're waving at you on the Google Hangout :) |
| 11:24 |
bshum |
via kmlussier's laptop |
| 11:25 |
|
yboston joined #evergreen |
| 11:25 |
phasefx |
wrong jason waved back :) |
| 11:25 |
mmorgan |
too many jasons! |
| 11:26 |
RoganH |
https://twitter.com/roganhamby/status/515160384370909184/photo/1 |
| 11:27 |
* kmlussier |
hopes nobody noticed her wild gesticulations as she was speaking negative balance issues with Dyrcona. :) |
| 11:27 |
DPearl |
kmlussier: You were out of frame. We missed it. |
| 11:27 |
kmlussier |
RoganH++ |
| 11:28 |
RoganH |
Whoever is sitting next to me ends up out of the picture frame. |
| 11:30 |
|
sandbergja joined #evergreen |
| 11:45 |
|
RoganH joined #evergreen |
| 11:46 |
kmlussier |
RoganH: I'll take a roast beef. |
| 11:46 |
RoganH |
kmlussier: mail it to you? |
| 11:46 |
RoganH |
kmlussier: parcel post? |
| 11:46 |
kmlussier |
RoganH: Sure! I bet it will be tasty by the time it gets here. |
| 11:47 |
RoganH |
kmlussier++ |
| 11:49 |
|
dMiller joined #evergreen |
| 12:02 |
|
nhilton joined #evergreen |
| 12:03 |
|
vanya joined #evergreen |
| 12:05 |
bshum |
dbs: I'm considering my working branch to fix update_pofiles to be a potential "solution" for https://bugs.launchpad.net/evergreen/+bug/1373582 |
| 12:05 |
pinesol_green |
Launchpad bug 1373582 in Evergreen "Outated translation in cs-CZ.po file" (affected: 1, heat: 6) [Undecided,Confirmed] |
| 12:05 |
bshum |
Not sure about how to deal with past versions right now |
| 12:05 |
bshum |
But I can at least make sure we fix it for 2.7.1 |
| 12:10 |
kmlussier |
bshum++ |
| 12:21 |
|
dMiller joined #evergreen |
| 12:38 |
vanya |
Dycrona : Hi! I wanted to ask you about the awesome box project |
| 12:39 |
vanya |
When the user tags a checked in item as awesome, does its rating go high based on the number of tags or is there a weight to each tag based on the person checking in? |
| 12:41 |
pastebot |
"eeevil" at 64.57.241.14 pasted "berick ... to whence should this be push'd" (21 lines) at http://paste.evergreen-ils.org/15 |
| 12:42 |
Dyrcona |
vanya: I believe the more times something is classified as awesome, the higher its rating becomes. |
| 12:43 |
Dyrcona |
vanya: That would be on the awesome box end of things. |
| 12:43 |
Dyrcona |
vanya: I got your email. Haven't had much time to respond today. |
| 12:44 |
csharp |
@who is awesome? |
| 12:44 |
pinesol_green |
book` is awesome. |
| 12:44 |
csharp |
@who is awesomer? |
| 12:44 |
pinesol_green |
Stompro is awesomer. |
| 12:44 |
csharp |
@who is the awesomeist? |
| 12:44 |
pinesol_green |
mnsri is the awesomeist. |
| 12:51 |
berick |
eeevil: pushed the circ bits to collab/berick/lp1198465-checkin-lost-overdue-fines-within-main-xact |
| 12:54 |
kmlussier |
I was just about the replicate/demonstrate some of my negative balance issues, but the above looks like work is being done. Should I be waiting to replicate/re-test those issues? |
| 12:54 |
vanya |
Dycrona: Thank you. I'm currently working on artificial intelligence for my honors project. Therefore, this project sounds very cool to me. |
| 12:55 |
vanya |
I wanted to know how the evergreen side of it works, and what all we need to install to work on it. |
| 12:56 |
vanya |
Also, I was wondering- what if we could train the system to give weightage to people's "awesome" tags on items, depending upon how much their previous tags are appreciated by other people. |
| 12:56 |
vanya |
(just an idea) |
| 12:57 |
RoganH |
For those in Rock Hill, Google Doc for placing orders for lunch tomorrow menu link included in doc. https://docs.google.com/document/d/1HxiQB4wJDyjBqFmIQHOHQf9d55hmOXi2lTL64F11JK4/edit?usp=sharing |
| 12:59 |
kmlussier |
My own personal opinion (others may have other opinions) is that, as a first step, that Evergreen sends the awesome info to Awesome Box using the API, because I think there will always be libraries that want to see their awesome titles along with the awesome titles from other libraries. |
| 13:00 |
kmlussier |
Beyond that first step, I think it would be great if Evergreen could do lots of awesome things with that data, and the idea you just raised sounds like it could be interesting. One question that immediately came to mind is how would you measure if other people appreciated that person's awesome tags? |
| 13:01 |
kmlussier |
But that's just one person's opinion. I would love to hear what others think. :) |
| 13:04 |
vanya |
kmlussier : Let's say you tag a book to be awesome. Now, if 100 other people check that book in, and 80 of them also tag it to be awesome- it will mean that your opinion matches a majority of people. |
| 13:04 |
vanya |
Now, if 100 other people check that book in and 5 of them tag it as awesome, your awesome tag is not in coherence with the majority. |
| 13:04 |
kmlussier |
Ah, I see. |
| 13:04 |
* kmlussier |
is intrigued. |
| 13:05 |
vanya |
Here, 80 and 5 are tentative numbers and can be changed according to user inputs. |
| 13:05 |
vanya |
So, in the former case, your awesome tag can be given more weightage as compared to the latter. |
| 13:06 |
* kmlussier |
wonders what eeevil thinks of these ideas and how they would play into potentially using the awesome factor in future activity metric ranking. |
| 13:08 |
|
cherri joined #evergreen |
| 13:08 |
vanya |
Let's say a book is tagged to be awesome by you. We can have a count of how many other people checked that book in, and how many of them tagged it to be awesome. The ratio of these two counts can provide the required support. |
| 13:09 |
vanya |
Also, the weightage may vary according to genres. So- you may have a good taste in mystery books but your taste in classical literature might not be the same as the majority crowd. |
| 13:09 |
kmlussier |
Part of me thinks it's a good idea. But I'm also thinking about people who don't typically read bestsellers. The things they mark as awesome aren't likely to get markes as often because they don't get checked out as often. Ah, but you just answered my question. It would be a ratio. |
| 13:10 |
vanya |
So- the weightage of your awesome tag in mystery would be higher than classical literature. |
| 13:11 |
vanya |
We can even extend it to provide recommendations to users depending on their coherence with other users with similar taste. |
| 13:12 |
vanya |
So- evergreen can provide an interpretive support to awesomebox. |
| 13:12 |
vanya |
(just a suggestion) |
| 13:14 |
kmlussier |
vanya: I like fact that you're thinking beyond the original project idea, and this all sounds good to me. This is also a good place to get feedback from the developers in the community. Many of them are at an event right now, so you might get more feedback from them later. |
| 13:15 |
|
dMiller joined #evergreen |
| 13:15 |
kmlussier |
vanya: But as you flesh out these idea, I think it might also be useful to put some of the ideas out to the Evergreen list. You're more likely to reach end users there, and it would be good to get their feedback. |
| 13:15 |
kmlussier |
It all sounds good to me, but I'm just one person. :) |
| 13:15 |
* eeevil |
reads up |
| 13:15 |
kmlussier |
vanya: When I say Evergreen list, I mean Evergreen general list. |
| 13:18 |
eeevil |
csharp: (scalar(@ou_ids) ? (owning_lib => \@ou_ids) : ()) |
| 13:19 |
vanya |
Thank you showing interest in my crazy ideas. Yes- you have a point. I'll send out the mail right away. |
| 13:19 |
vanya |
*for |
| 13:20 |
kmlussier |
vanya: Stick around here for a while too in case any of the other devs have any feedback. |
| 13:20 |
vanya |
Sure :) |
| 13:21 |
kmlussier |
Or I should say any of the devs since I'm not a developer. :) |
| 13:25 |
berick |
dbwells: collab/berick/lp1198465-checkin-lost-overdue-fines-within-main-xact |
| 13:42 |
* Dyrcona |
is mostly not paying attention to IRC. |
| 13:43 |
Dyrcona |
vanya: I'll answer some of your questions in email and see if I can find some ideas that some of us already discusses regarding the awesome box idea. |
| 13:45 |
vanya |
Dyrcona: Okay |
| 13:52 |
|
RoganH joined #evergreen |
| 13:52 |
|
snigdha26 joined #evergreen |
| 13:53 |
eeevil |
vanya: re kmlussier's poke of me, and the activity metric, I think there's definitely some overlap, but what you're talking about is much more patron-specific (identifying preference coherence), where the other project focuses mainly on identifying the aggregate "popularity" etc. so, really the infrastructure created for the activity metric project might be a data source for what you're thinking |
| 13:55 |
eeevil |
vanya / kmlussier: IOW, the awesomebox tag is something we've talked about collecting and feeding into an aggregate popularity rank for a title. going back and doing a second-order analysis of the per-patron tagging sounds really cool, though, and could definitely be a starting point for a recommendation engine |
| 13:56 |
kmlussier |
eeevil: But it might be likely that activity metric infrastructure isn't built by the time the OPW projects begin. In that case, I'm wondering if there is a particular approach she should be using to make it easier for future development where her awesome factors can be used as a popularity metric. |
| 13:58 |
csharp |
webby++ |
| 13:58 |
* csharp |
wants a duck logo |
| 13:58 |
berick |
clippy-- |
| 13:59 |
kmlussier |
csharp: To go with the turtle? |
| 13:59 |
vanya |
kmlussier : a database would be needed as a training set, in order to build a recommendation system, which, as you pointed out, might not be available by the time OPW projects begin. So, I could begin with a dummy database and use it for my work. And later, when the project is up and running, we can directly use it as a black box on the real data. |
| 13:59 |
vanya |
What do you think? |
| 14:00 |
dbs |
vanya: the ideas in your email sound awesome |
| 14:01 |
* dbs |
isn't quite sure if by "a database" you mean just another table in the existing Evergreen database, or an entirely separate database |
| 14:04 |
vanya |
dbs : thank you :) Any suggestions for improvement? Sorry for the ambiguity, by database, I meant a record of users and their respective awesome tags, and the number of people who checked in the book and those among them who marked it awesome. |
| 14:04 |
|
dMiller joined #evergreen |
| 14:04 |
RoganH |
bshum: I was curious and looked at that instance you mentioned and it doesn't violate the trademark policy |
| 14:05 |
vanya |
So, we might say it is a table that looks like: user | book tagged awesome | number of other users who checked that book in | number of awesome tags by other users |
| 14:14 |
|
Shae joined #evergreen |
| 14:27 |
kmlussier |
vanya: For now, I would proceed as if that other activity metric project wasn't happening with a real table that stores the data. As we get closer to the OPW dates, we should have more information on whether we've secured funding for that other project and, if so, when it might be happening. |
| 14:33 |
vanya |
kmlussier : since it is on similar lines as the awesome box project, what if, instead of treating it as a separate project, we take it as an add on to the original awesome box project? |
| 14:38 |
kmlussier |
vanya: There is some overlap, but I think eeevil is correct when he says they are distinct things. Awesome tags were just one very small factor being considered for that project. So I wouldn't worry about it now. |
| 14:39 |
kmlussier |
graced++ # Making us discuss difficult questions. |
| 14:39 |
bshum |
eeevil++ # speedy database! |
| 14:39 |
graced |
kmlussier++ for reading the tome :) |
| 14:39 |
bshum |
magic SQL! |
| 14:44 |
kmlussier |
hmmm...I'm now pondering the question on whether there is a way that awesome data could be anonymized in some way based on the concern voiced in Terran's e-mail. |
| 14:51 |
|
jboyer_laptaupe joined #evergreen |
| 14:51 |
|
dbwells__ joined #evergreen |
| 14:53 |
eeevil |
magic troubleshooting and investigation and reasoning ... but I digress ;) |
| 14:53 |
bshum |
eeevil: https://bugs.launchpad.net/evergreen/+bug/1374091 |
| 14:53 |
pinesol_green |
Launchpad bug 1374091 in Evergreen "unapi.bre and slow views" (affected: 1, heat: 6) [Undecided,New] |
| 14:53 |
eeevil |
bshum: thanks |
| 14:53 |
vanya |
kmlussier : I couldn't exactly get how the privacy issues occur. Could you please elaborate on that? |
| 14:55 |
kmlussier |
vanya: As a profession, librarians try not to track the materials specific people check out beyond a certain time period. Or I should say we don't encourage. |
| 14:55 |
|
dbwells joined #evergreen |
| 14:55 |
kmlussier |
So in Evergreen, we do have a method set up where we can archive circulation transactions after a certain period of time. Those archived transactions are available for statistical purposes, but they are no longer linked back to a particular user. |
| 14:56 |
kmlussier |
I don't know how many sites actually anonymize their transactions, but patron privacy is something we try to maintain. |
| 14:57 |
kmlussier |
At the same time, there are cases where a historical record of a user's checkouts are maintained. For example, some users like to keep a history of the materials they have checked out. In those cases, then, we are not anonymizing those transactions. |
| 14:58 |
kmlussier |
But, in that case, the users have chosen to opt into having that history kept. |
| 14:59 |
vanya |
kmlussier : Thank you for explaining. In this case, we are not actually maintaining the record of the user's transactions. All I'm proposing to retain is the ratio of the two counts. Will that be a problem too? |
| 15:00 |
jcamins |
vanya: how would you know to increment the counts? |
| 15:01 |
kmlussier |
vanya: Sure, but that ratio informs the system how important a particular user's awesome rating is, right? So it seems like somewhere along the way, you would need to link a particular user with a particular title. |
| 15:01 |
kmlussier |
jcamins: Increment the awesome counts/ |
| 15:01 |
kmlussier |
? |
| 15:04 |
jcamins |
kmlussier: the "good prediction" and "not good prediction" counts. |
| 15:04 |
* jcamins |
didn't see the mailing list, just read the description on IRC earlier and thought it sounded incredibly awesome. |
| 15:04 |
kmlussier |
jcamins: ah, ok. I'll defer to vanya then. |
| 15:04 |
kmlussier |
How many times can we use the word "awesome" to describe the awesome box project? :) |
| 15:05 |
jcamins |
kmlussier: a lot! |
| 15:05 |
phasefx |
kmlussier: I think the simplest case is to think of it as two circ count tallies.. one for awesome checkins, and the other for non-awesome checkins.. no patrons required.. but also harder to deprecate or age the score.. it becomes a lifetime ratio for awesomeness |
| 15:06 |
|
jboyer-laptaupe joined #evergreen |
| 15:07 |
phasefx |
unless I'm missing how the whole "second-order analysis" part works |
| 15:07 |
kmlussier |
phasefx: Well, you could add the date something was made awesome without compromising patron privacy, right? So that would take care of deprecation. |
| 15:07 |
phasefx |
kmlussier: yeah, surmountable, just more work |
| 15:08 |
phasefx |
with dates, you still have some fuzzy linkages big brother could use in certain circumstances |
| 15:09 |
dbs |
Vanya's artificial intelligence "users like you also liked X" will suffer, the more the data is anonymized |
| 15:10 |
* dbs |
worked a little bit on such an algorithm at https://github.com/splurge/splurge |
| 15:11 |
jcamins |
dbs: right, that's the incrementing I was wondering about. |
| 15:12 |
dbs |
So the anonymity could be optional, just like our current transaction data is optional |
| 15:12 |
vanya |
as kmlussier mentioned, some users allow the library to maintain the records, we can build the algorithm using the data of those people. |
| 15:13 |
kmlussier |
dbs: I was thinking it might be workable to have users opt into it. But, then, how many patrons would actually take the time to do it? |
| 15:13 |
dbs |
If you wanted to make it really complicated, users also opt into storing their checkout history -- but yeah, my guess is not many take the time to do that. |
| 15:14 |
* kmlussier |
wonders what the percentage of patrons opt in now. |
| 15:14 |
kmlussier |
Percentage of active patrons. Not percentage of all patrons. |
| 15:14 |
jcamins |
kmlussier: if only you had an ILS with solid reporting capabilities! |
| 15:14 |
* phasefx |
bets low |
| 15:14 |
kmlussier |
jcamins: Sure, but I don't know how run those reports. :) |
| 15:15 |
kmlussier |
Which is why I wonder aloud in IRC. With the hopes that somebody will hear me. |
| 15:15 |
vanya |
kmlussier : you do have a point that not many patrons would directly agree to let their transactions be recorded. But as a tradeoff, they will have a better experience with the library. |
| 15:15 |
jcamins |
If only you knew how to find people who were experts on the ILS's reporting capabilities! |
| 15:15 |
phasefx |
conversely, I bet awesome box usage is high where it gets used, since it's so easy to participlate |
| 15:15 |
collum |
The first difficult thing would be to define 'active' |
| 15:15 |
vanya |
For example, certain websites that offer to record searches in order to be able to provide recommendations. |
| 15:15 |
kmlussier |
collum: Somebody who has checked out items in the past year? |
| 15:16 |
collum |
or logged into a computer, or used a database, or used an ebook |
| 15:16 |
vanya |
The user mainly has a tradeoff between privacy and user experience. |
| 15:16 |
kmlussier |
vanya: I think the difference here is that the place where the patron is when they return an awesome item isn't the same place where they mark their perference (on a computer). |
| 15:18 |
kmlussier |
collum: I don't think logging into a computer or using a database is relevant in this case. Because they don't have a circ history to record. |
| 15:18 |
jcamins |
collum: you could probably simplify it to "checked out an item in the last year," because saving the checkout history of anyone who hasn't checked out any item in the last year would be, practically speaking, meaningless. |
| 15:19 |
dbs |
37 total in our entire consortium |
| 15:19 |
collum |
But these folks would potentially check out something. |
| 15:19 |
dbs |
SELECT COUNT(*) FROM actor.usr_setting WHERE name = 'history.circ.retention_start'; |
| 15:20 |
phasefx |
congratulatoins, you 37 are now curators |
| 15:20 |
dbs |
vanya: yep, that's the classic core tradeoff |
| 15:21 |
dbs |
Maybe part of the awesome box rollout could be a campaign to advertise the circ history setting. And make it a one-click opt-in. |
| 15:22 |
dbs |
That could be entirely separate from vanya's awesome box implementation, of course. |
| 15:22 |
collum |
OK. Reading bath through more closely. I'll concede. :-) |
| 15:23 |
collum |
s/back/bath - oy |
| 15:24 |
vanya |
dbs : Yes that could be the only way around the privacy issues. But in general, these days every wants user friendly environment and I guess a lot would not mind opting in. |
| 15:33 |
dbs |
vanya: yes, if they understand the benefits then they might take the time. Gotta keep Privacy by Design in mind. |
| 15:33 |
kmlussier |
OK, as I said before, I'm not a developer, so I may be saying things that don't make sense. If all of this awesome data is in its own table, then we have the problem of needing to check that user's setting to see if the data should be stored. |
| 15:34 |
dbs |
That's not a big problem |
| 15:34 |
jcamins |
Of course, it strikes me that encouraging people to disable their privacy settings would compromise *other* people's privacy, too. At least potentially. |
| 15:34 |
kmlussier |
However, if it is somehow it is linked to data in the action.circulation table, then the amount of data you get all depends on whether a) the Evergreen site is even archiving circulations and b) if the user has opted in. |
| 15:35 |
kmlussier |
So then the awesome tag isn't adding more data about the user's circulation history. It's using data that's already there. |
| 15:35 |
kmlussier |
Am I making any sense? |
| 15:36 |
jcamins |
A fishing expedition through the ILS records would turn up the N users in town who might have something to hide (or might not see any use to changing their setting). |
| 15:36 |
dbs |
kmlussier: I'm imagining that we could handle all cases, and that the results are simply richer if a user has opted in |
| 15:36 |
jcamins |
That's a policy issue not a technical issue, but it's an interesting thought. Sort of like Tor isn't very useful if only spies are using it. |
| 15:36 |
dbs |
jcamins: sure, that could happen with or without an ILS |
| 15:37 |
dbs |
Someone could be standing next to the circ desk watching what people are checking out too. |
| 15:37 |
dbs |
kmlussier: as phasefx said, in the worst case we have the anonymized user (user id = 1) accounting for all of the awesome / not awesome tallies, so it's a generalized set of data. |
| 15:38 |
* dbs |
returns to strike preparation duties |
| 15:38 |
kmlussier |
dbs: And then the non-anonymized data builds on top of it and enriches it? OK, I can go with that. |
| 15:39 |
jcamins |
dbs: right, but I have no expectation that when I go into the library I'll be protected from surveillance. Whereas "we try to resist the patriot act" doesn't mean much if you're giving up enough data that no additional data mining is needed. |
| 15:40 |
jcamins |
(or whatever it is the ambassadors at my undergrad institution were saying when I was there) |
| 15:41 |
dbs |
I think the one thing we need in the checkout history is the ability for the user to delete individual items from their history, if they so desire. (although we're running 2.4, maye that has been added already). |
| 15:42 |
kmlussier |
dbs: DPearl is working on something like that. |
| 15:42 |
kmlussier |
I think. |
| 15:42 |
dbs |
win 27 |
| 15:42 |
dbs |
lose it all |
| 15:43 |
kmlussier |
bug 1312699 |
| 15:43 |
pinesol_green |
Launchpad bug 1312699 in Evergreen "Editable Checkout History" (affected: 5, heat: 22) [Wishlist,Triaged] https://launchpad.net/bugs/1312699 - Assigned to Dan Pearl (dpearl) |
| 15:43 |
kmlussier |
dbs: Wait, 2.4? Wow! |
| 15:44 |
dbs |
kmlussier: yeah, tell me about it. |
| 15:44 |
vanya |
dbs : That is an awesome point! We can simplify the privacy issues to a great extent by letting them delete whichsoever entries they want off record. |
| 16:18 |
kmlussier |
A question for the OPW mentors. Were you all added to any opw mailing lists? I just need to know if I should be forwarding communication to you. |
| 16:20 |
dbs |
I don't think I was added to an OPW mailing list. |
| 16:21 |
dbs |
Now I'm wondering if I forgot to do something. |
| 16:21 |
kmlussier |
dbs: I was automatically added, so I thought you may have been too. But I'll send the info to everyone in case you want to be added to the coordinators/mentors list. |
| 16:21 |
kmlussier |
Glad I asked the question. :) |
| 16:22 |
|
gsams joined #evergreen |
| 16:22 |
* kmlussier |
really needs to write a blog post. |
| 16:24 |
* kmlussier |
really needs to remove the "volunteer" word from her vocabulary. |
| 16:27 |
|
tspindler left #evergreen |
| 16:28 |
dbs |
kmlussier: attach weights to your wrists to keep your hands down. |
| 16:29 |
kmlussier |
Well, then, if I did volunteer, at least I would be giving my arms a workout at the same time. :) |
| 16:30 |
kmlussier |
I generally enjoy volunteering for Evergreen things. Volunteering to be GS cookie mom, though, has left me questioning my sanity. |
| 16:30 |
|
vanya joined #evergreen |
| 16:31 |
dbs |
kmlussier: That would leave me about 50 pounds heavier. |
| 16:33 |
jcamins |
kmlussier: yay, cookies! |
| 16:34 |
jcamins |
Oh, wait, the cookies come pre-baked. Never mind. |
| 16:34 |
kmlussier |
jcamins: But I won't be baking them this time. |
| 16:34 |
kmlussier |
Yeah, that. |
| 16:39 |
* kmlussier |
sometimes wonders if jcamins has an alert set up to tell him whenever somebody says the word "cookie" in an IRC channel. |
| 16:46 |
kmlussier |
@dessert [someone] |
| 16:46 |
* pinesol_green |
grabs some Lemon Chess Pie for vanya |
| 16:46 |
csharp |
nom nom nom |
| 16:46 |
|
RoganH joined #evergreen |
| 16:46 |
kmlussier |
Ah nice! pinesol_green should always give treats to new people in the community. |
| 16:46 |
vanya |
did someone say treat? |
| 16:47 |
kmlussier |
vanya: Yes, pinesol_green is our channel bot, and we have her programmed to give desserts to people when we use the @dessert command. |
| 16:49 |
vanya |
Wow! That's so cool! |
| 16:49 |
vanya |
Can I try? |
| 16:49 |
kmlussier |
Absolutely! |
| 16:49 |
vanya |
@dessert [someone] |
| 16:49 |
* pinesol_green |
grabs some Lemon Chess Pie for DPearl |
| 16:49 |
kmlussier |
You can also direct it to a specific person. |
| 16:49 |
vanya |
Wow |
| 16:49 |
kmlussier |
@dessert csharp |
| 16:49 |
* pinesol_green |
grabs some of mllewellyn's Cupcakes for csharp |
| 16:50 |
kmlussier |
Ooh! Lucky csharp. |
| 16:50 |
* phasefx |
salivates |
| 16:50 |
vanya |
@dessert kmlussier |
| 16:50 |
* pinesol_green |
grabs some Chocolate Mousse for kmlussier |
| 16:50 |
vanya |
pinesol_green is making me hungry |
| 16:50 |
pinesol_green |
vanya: I am only a bot, please don't think I'm intelligent :) |
| 16:50 |
pinesol_green |
vanya: Sorry, that command is only available to Evergreen Premium™ Subscribers. Please upgrade your subscription ASAP! |
| 16:51 |
vanya |
It talks too :o |
| 16:51 |
vanya |
Amazing! |
| 16:51 |
vanya |
Who wrote it? |
| 16:52 |
kmlussier |
Hmmm...it's a plugin based on something used in another IRC channel. I know there's a github branch out there. Let me see if I can find it. |
| 16:52 |
phasefx |
vanya: http://sourceforge.net/projects/supybot/ |
| 16:54 |
kmlussier |
Ah, okay. So https://github.com/code4lib/supybot-plugins/ is a plug-in I see. |
| 17:00 |
|
cherri joined #evergreen |
| 17:10 |
|
mmorgan left #evergreen |
| 17:17 |
pinesol_green |
[evergreen|Chris Sharp] LP#1373693: Speed up call number browse search at top level org. - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=63a2667> |
| 17:20 |
bshum |
csharp++ |
| 17:25 |
pinesol_green |
[evergreen|Mike Rylander] LP#1347774 Anonymous PCRUD mode - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=aa2931c> |
| 17:25 |
pinesol_green |
[evergreen|Bill Erickson] LP#1347774 CStoreEditor anonymous PCRUD additions - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=951b648> |
| 17:28 |
pinesol_green |
Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html> |
| 17:28 |
|
whargrove joined #evergreen |
| 17:28 |
|
yboston_ joined #evergreen |
| 17:29 |
|
yboston__ joined #evergreen |
| 17:30 |
whargrove |
Hi all, I'm working on SIP2 integration with evergreen and we're getting an error when sending the patron information request ... Use of uninitialized value in subroutine entry at /usr/lib/perl/5.14/Encode.pm line 202, <STDIN> chunk 1. |
| 17:30 |
whargrove |
Any ideas what that could mean? |
| 17:32 |
whargrove |
Our request looks like this: "6300020140925 021844 AOgapines|AAa|ADb\r" |
| 17:36 |
|
yboston joined #evergreen |
| 17:57 |
csharp |
@quote add <bshum> we should do something about those cupcakes - they're going to go to waste |
| 17:57 |
pinesol_green |
csharp: The operation succeeded. Quote #94 added. |
| 18:21 |
|
mrpeters left #evergreen |
| 18:29 |
|
DPearl joined #evergreen |
| 18:35 |
|
Dyrcona joined #evergreen |
| 19:02 |
|
dbwells__ joined #evergreen |
| 19:17 |
|
kmlussier joined #evergreen |
| 20:02 |
|
mrpeters joined #evergreen |
| 20:10 |
|
kmlussier joined #evergreen |
| 20:16 |
kmlussier |
@marc 655 |
| 20:16 |
pinesol_green |
kmlussier: Terms indicating the genre, form, and/or physical characteristics of the materials being described. A genre term designates the style or technique of the intellectual content of textual materials or, for graphic materials, aspects such as vantage point, intended purpose, or method of representation. A form term designates historically and functionally specific kinds of materials (2 more messages) |
| 21:23 |
|
yboston joined #evergreen |
| 21:44 |
|
dbwells joined #evergreen |
| 21:45 |
|
dbwells_ joined #evergreen |
| 21:51 |
|
dbwells joined #evergreen |
| 21:54 |
|
dbwells_ joined #evergreen |
| 21:56 |
|
jboyer-laptaupe joined #evergreen |
| 22:02 |
|
dbwells joined #evergreen |
| 22:23 |
|
DPearl joined #evergreen |
| 22:32 |
* jeff |
yawns |
| 22:51 |
|
jventuro joined #evergreen |
| 23:29 |
|
cherri joined #evergreen |
